Neuro-symbolic AI for opinion mining

Summarise with VIRAT
Authors: Zheng Wei Liau
Cited by: 0

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

Over the past few years, Artificial Intelligence has grown tremendously and has become a part of our everyday lives. With its extensive problem-solving abilities, it is now being incorporated into decision making processes, whether it is in the workplace or our daily lives. On the surface, AI seems like the ultimate tool that can deal with any problem. However, it has been known to hallucinate and produce strange or biased answers. To tackle this conundrum, the SenticNet team has developed a neurosymbolic AI model that serves to explain its reasoning processes while still being able to handle comprehensive tasks. Still, this model is far from perfect and requires further fine tuning. As such, to understand the capabilities of the SenticNet AI model, a dataset of social media text, particularly tweets, concerning pollution has been developed. This allows for greater insight into how a dataset can be developed and automated for the SenticNet AI model in the future. On top of that, we can understand any areas whereby SenticNet AI model has underperformed. After which, a multitude of solutions are suggested. These solutions would breathe new life into developing a superintelligent AI model that is highly accurate and can explain its processes without hallucination. Additionally, a slew of recommendations is listed which can be undertaken for future work on the SenticNet AI model.

Fields of Study & Subjects

Sentiment Analysis and Opinion Mining Topic Modeling Ethics and Social Impacts of AI

Publication details

Journal / source
DR-NTU (Nanyang Technological University)
Publication date
1 January 2026
Language
EN